Claude Opus 5 (Adaptive Reasoning, Max Effort): benchmark results
Provider: Anthropic. Released 2026-07-24. Access: API.
Unified ELO 1744 ± 1, rank #38 of 1761 rated models, from 25 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Artificial Analysis Intelligence Index | 54.05 | Intelligence Index | 99.5 |
| AA Humanity's Last Exam | 54.87 | Accuracy (%) | 99.3 |
| AA GDPval | 1739.24 | ELO | 99.2 |
| AA Omniscience - Business | 53 | Accuracy (%) | 99.2 |
| AA Omniscience - Science, Engineering & Mathematics | 57.52 | Accuracy (%) | 99 |
| AA Omniscience - Humanities & Social Sciences | 58.6 | Accuracy (%) | 98.4 |
| AA-Omniscience Accuracy | 60.87 | Accuracy (%) | 98.4 |
| AA CritPt | 29.14 | Accuracy (%) | 98.1 |
| AA Omniscience | 37.07 | Score | 98.1 |
| AA Omniscience - Software Engineering (SWE) | 86 | Accuracy (%) | 97.9 |
| AA Omniscience - Health | 52.27 | Accuracy (%) | 97.7 |
| Riemann-bench | 68 | Score (%) | 97.3 |
Interactive version: theaggregate.ai/model?slug=claude-opus-5-adaptive-reasoning-max-effort · How It Works · Data refreshed daily, snapshot 2026-09-05.